Method and system for processing MRT data of the human brain

1. A method for processing MRT data of a brain of a patient, wherein three-dimensional MRT data resolved in voxels of the brain of the patient and brains of a normative database of a plurality of neurologically healthy human individuals are available, comprising the steps of: segmenting the MRT data of the brain of the patient for each voxel into portions of grey substance, white substance and liquor by way of a classification algorithm, transforming the data to be compared with corresponding data of the normative database into a standardized stereotactic space, resulting in one or more normalized tissue maps that contain proportions determined by the classification algorithm, wherein the transformation occurs before, after or synchronous with segmentation, and subjecting the tissue map or tissue maps of the brain of the patient to a voxel-wise statistical comparison with correspondingly normalized and segmented data of the normative database, wherein the voxel-wise statistical comparison of the tissue map, or tissue maps, is a morphometric comparison, wherein statistically significant voxels in the tissue map, or tissue maps, of the brain of the patient are identified that belong to a region dominated by the white substance that contains more grey substance than should be expected from the normative database, or that belong to a region dominated by the grey substance that contains more white substance than would be expected from the normative database.

2. The method according to claim 1 , wherein, in addition, such statistically significant voxels are identified in the tissue map, or tissue maps, of the brain of the patient that belong to a region dominated by the white substance in which less white substance is contained than is to be expected from the normative database, or that belong to a region dominated by the grey substance in which less grey substance is contained than is to be expected from the normative database.

3. The method according to claim 1 , wherein coherent clusters of statistically significant voxels are measured and/or counted and/or depicted as a histogram with reference to their volume.

4. The method according to claim 1 , wherein the segmented MRT data is smoothed to create the tissue map or tissue maps.

5. The method according to claim 1 , wherein patterns arising from the voxel-wise statistical comparison are imaged.

6. The method according to claim 1 , wherein voxels of the tissue map, or tissue maps, are interpolated into regular or irregular grids.

7. The method according to claim 1 , wherein parameters of the classification algorithm are iteratively adjusted to reproduce published results for the MRT data of the normative database and/or finely adjusted to a ground truth of simulated data, wherein the classification algorithm for the voxel-wise statistical comparison is used with same parameters for the MRT data from the normative database as well as from the brain of the patient.

8. The method according to claim 1 , wherein one or more volumetric tests are performed additionally in which volumes of white and grey substance can be calculated from the tissue maps using masks for regions of the brain of the patient to be examined.

9. The method according to claim 8 , wherein the volumes are adapted to individual data of the patient by way of a correction with respect to a covariate, or a plurality of covariates, including head size, age, duration of illness and/or gender, and/or are standardized to an overall brain substance, overall white or grey substance, or overall intracranial volume.

10. The method according to claim 9 , wherein the adapted volumes are corrected for age.

11. The method according to claim 8 , wherein the volumes are statistically compared against the normative database and/or depicted as graphs.

12. The method according to claim 1 , wherein an additional test is performed to determine tissue damage wherein a three-dimensional mask is prepared in whose region the white or grey substance predominates, a value of the tissue maps are added separately in each case for the white and grey substance of the patient and data of the normative database for the region of the mask, and thus volumetric volumes for an overall grey and white substance are thereby obtained within the mask for the patient and normative database in each case, and these are brought into relation with each other.

13. The method according to claim 12 , wherein the tissue damage is a lesion load.

14. The method according to claim 1 , wherein said method is used for evaluating hypointense lesions in the white substance as depicted in T1-weighted data for multiple sclerosis patients.

15. A non-transitory computer readable medium stored with a computer program, that when executed by a computer, causes the computer to process MRT data of a brain of a patient, wherein three-dimensional MRT data resolved in voxels of the brain of the patient and brains of a normative database of a plurality of healthy human individuals are available, and when the computer program is run, performs steps comprising: segmenting the MRT data of the brain of the patient for each voxel into portions of grey substance, white substance and liquor by way of a classification algorithm, transforming the data to be compared with corresponding data of the normative database into a standardized stereotactic space, resulting in one or more normalized tissue maps that contain proportions determined by the classification algorithm, wherein the transformation occurs before, after or synchronous with segmentation, and subjecting the tissue map or tissue maps of the brain of the patient to a voxel-wise statistical comparison with correspondingly normalized and segmented data of the normative database, wherein the voxel-wise statistical comparison of the tissue map, or tissue maps, is a morphometric comparison, wherein statistically significant voxels in the tissue map, or tissue maps, of the brain of the patient are identified that belong to a region dominated by the white substance that contains more grey substance than should be expected from the normative database, or that belong to a region dominated by the grey substance that contains more white substance than would be expected from the normative database.

16. The non-transitory computer readable medium stored with a computer program according to claim 15 , wherein, in addition to the classification algorithm, a normalization algorithm and/or a smoothing algorithm is or are implemented.

17. A system for processing MRT data of a brain of a patient, wherein three-dimensional MRT data resolved in voxels of the brain of the patient and brains of a normative database of a plurality of healthy human individuals are available, comprising a data processing system having a memory for the MRT data of the brain of the patient, and the normative database, and for tissue maps on which non-transitory computer readable medium stored with a computer program that is programmed to: segment the MRT data of the brain of the patient for each voxel into portions of grey substance, white substance and liquor by way of a classification algorithm, transform the data to be compared with corresponding data of the normative database into a standardized stereotactic space, resulting in one or more normalized tissue maps that contain proportions determined by the classification algorithm, wherein the transformation occurs before, after or synchronous with segmentation, and subject the tissue map or tissue maps of the brain of the patient to a voxel-wise statistical comparison with correspondingly normalized and segmented data of the normative database, wherein the voxel-wise statistical comparison of the tissue map, or tissue maps, is a morphometric comparison, wherein statistically significant voxels in the tissue map, or tissue maps, of the brain of the patient are identified that belong to a region dominated by the white substance that contains more grey substance than should be expected from the normative database, or that belong to a region dominated by the grey substance that contains more white substance than would be expected from the normative database.